AiGpu

Industry·

Signal Restricts Screenshots to Block On‑Device AI Agents From Reading Chats

Signal is disabling screenshot capture on iOS and Android to stop system‑level AI assistants such as Siri and Gemini from accessing encrypted message content, calling the current OS integration a privacy threat.

Signal app privacy settings screen showing screenshot block toggle

Signal has introduced a new privacy control that blocks the ability to take screenshots inside its mobile apps on iOS 27 and Android. The move is aimed directly at on‑device AI agents — Apple’s Siri, Google’s Gemini, and similar assistants — that can scan on‑screen content to build user profiles or feed data into larger models.

Users can re‑enable screenshots from the app’s settings, but Signal warns that doing so “potentially exposes your (and your contacts’) Signal content to these AI agents, and breaks our privacy guarantees.” The feature mirrors a similar restriction the company added last year to thwart Microsoft’s Recall feature.

Why it matters for AI infrastructure

  • On‑device inference increasingly relies on GPU‑accelerated pipelines; limiting screen access reduces the volume of sensitive data that could be harvested for training or fine‑tuning.
  • Developers of AI‑enabled services must now account for stricter app‑level boundaries when designing data‑collection workflows.
  • Hardware vendors supplying AI accelerators for mobile SoCs should anticipate tighter OS‑level sandboxing as a default.

Signal President Meredith Whittaker described the current OS integration of AI agents as “hostile to privacy” and called on Apple, Google, and Microsoft to provide simple technical tools and accurate documentation so that app developers can enforce data‑access policies without sacrificing core user experience. The company acknowledges that screenshot blocking is an incomplete fix, as agents may still reach data through filesystem APIs or deep accessibility hooks.

Until platform providers expose robust, developer‑controlled safeguards, encrypted‑messaging services and other privacy‑focused apps will continue to rely on blunt instruments like screenshot bans. The episode underscores a growing tension between on‑device AI convenience and the confidentiality guarantees that enterprises and end‑users expect from their communication tools.

  • aigpu
  • ai gpu
  • ai gpu cloud
  • aigpu dubai
  • privacy
  • signal
  • on-device ai
  • mobile ai

By AiGpu Editorial · Editorial rewrite based on public reporting (The Verge AI)

← All articles